We compared four processes that activate cell proliferation in the liver: (1) hyperplasia induced by the tumor promoter, TCPOBOP;(2) antipromotional hyperplasia induced by thyroid hormone;(3) rapid liver regeneration after partial hepatectomy (PH) in the rat;and (4) more protracted regeneration after PH in the mouse. The first phase of the project generated a comprehensive system of microarray expression profiles of all four processes. TCPOBOP is a direct ligand for the nuclear receptor transcription factor, CAR, and rapidly induces liver growth and cell proliferation. The expression profiles showed rapid induction of Gadd45[unreadable], Jun, and Fosl2 mRNA after TCPOBOP treatment;we focused on transcriptional regulation of these three genes, and on Cyclin D1 as their downstream target. A Gadd45[unreadable]-null mouse had active proliferation after TCPOBOP treatment, but impaired transcription and liver growth. Molecular studies showed that Gadd45[unreadable] directly binds CAR and acts as a transcriptional coactivator. Thus, rapid induction of Gadd45[unreadable] is anabolic and enables the high rate of transcription needed for rapid growth. However, other investigators have found that Gadd45[unreadable] regulates apoptosis and proliferation. The research proposed in Aim 1 will investigate how Gadd45[unreadable], only 18 kD, can function as a coactivator, by determining its binding partners in the transcription complex, and its effects on progressive chromatin remodeling during transcriptional activation. Other experiments will determine how Gadd45[unreadable] can have so many critical functions, by mutating individual protein domains and testing each for effects on CAR-binding, coactivation, activation, apoptosis, and proliferation. Additional studies will use adenovirus to reconstitute mutated Gadd45[unreadable] in the livers of null mice and test the importance of each separate function in hyperplasia. Aim 2 will address the diverse mechanisms by which CAR, a cis acting transcription factor, activates CAR- responsive regulatory regions of Fosl2. Experiments will also test whether Jun and Fosl2 are responsible for inducing hyperplasia, by studying TCPOBOP treatment in conditional Fosl2- and Jun-null mice, and by using adenovirus to induce liver expression vivo. Preliminary studies showed that Fosl2 and Jun combine to stimulate transcription of Cyclin D1 through a novel upstream regulatory region. Aim 3 will investigate the mechanisms that activate Cyclin D1 transcription and explain context-specific stimulation by Fosl2. Experiments will also determine the alternative pathways by which liver regeneration and T3-induced hyperplasia activate Cyclin D1 transcription. PUBLIC HEALTH RELEVANCE: The normal liver has several kinds of growth responses-some promote cancer while others inhibit promotion. These growth responses are very complex because they activate expression of hundreds of genes. The goals of this project are to define the common and distinctive parts of each response, and working from this comparison, determine a unique pathway of drug-induced proliferation and cancer promotion.